[Orca-users] Orca problems on Solaris 8 with SAN support

[Buhler Justin]

All, 

I spent the last week with Adrian C looking at the segmentation fault , which seems to happen to some people using Fiber Channel disks (ssd), MPxIO, or Sun Cluster. We found that the problem was with the live_rules.se and diskinfo.se classes found in SEtoolkit versions < 3.3.1. Once we upgrade the Setoolkit to 3.3.1 the problem went aways. A good way to test is to run the following command. 

se disks.se

The following is some output from a machine using 3.3 and not 3.3.1. You can see that the ssd drives are not being mapped correctly with device files in /dev/dsk

<<< VOLUME INFO >>>

kernel -> path_to_inst -> /dev/dsk part_count [fstype mount]

ssd0 -> ssd0 -> ssd0 0

ssd0,a -> ssd0,a -> ssd0,a 0

ssd0,c -> ssd0,c -> ssd0,c 0

ssd0,h -> ssd0,h -> ssd0,h 0

ssd1 -> ssd1 -> ssd1 0

ssd1,c -> ssd1,c -> ssd1,c 0

ssd1,d -> ssd1,d -> ssd1,d 0

ssd1,e -> ssd1,e -> ssd1,e 0

The following is the same run with se 3.3.1 and is correctly mapped. 

<<< VOLUME INFO >>>

kernel -> path_to_inst -> /dev/dsk part_count [fstype mount]

ssd1 -> ssd1 -> c1t1d0 0

ssd1,a -> ssd1,a -> c1t1d0s0 0

ssd1,b -> ssd1,b -> c1t1d0s1 0

ssd1,c -> ssd1,c -> c1t1d0s2 0

ssd1,d -> ssd1,d -> c1t1d0s3 0

ssd1,e -> ssd1,e -> c1t1d0s4 0

ssd1,h -> ssd1,h -> c1t1d0s7 0

I just thought that I would post this since I have seen this problem posted several time. 

Good Luck 

-----Original Message-----
From: Richard.Grevis at ubs.com [mailto:Richard.Grevis at ubs.com]
Sent: Monday, April 26, 2004 12:51 PM
To: david_ross at transcanada.com; orca-users at orcaware.com
Subject: RE: [Orca-users] Orca problems on Solaris 8 with SAN support


We use SAN with sun low level SAN related drivers and veritas. there is no problem, but then the names are not weird. Have you tried increasing the size of the appropriate declared arrays in the toolkiit source?
 
regards,
richard

[Grevis, Richard]  -----Original Message-----
From: orca-users-bounces+richard.grevis=ubs.com at orcaware.com [mailto:orca-users-bounces+richard.grevis=ubs.com at orcaware.com]On Behalf Of David Ross
Sent: 23 April 2004 20:28
To: orca-users at orcaware.com
Subject: [Orca-users] Orca problems on Solaris 8 with SAN support



We're trying to use Orca 0.27, with orcallator.se v1.37 on Solaris 8, using SE Toolkit 3.3.
We have it running on about 100 servers right now, no problem, except for one server we have now built with Sun's SAN Foundation Kit installed. 
 
Orcallator will not start - it segmentation faults.  Running SE Toolkit with th "-d" option shows this:
---
if (j<75> > <0>)
if (strncmp(first_name<ssd9>, short_name<md100>, first_len<4>) == <0>)
if (strcmp(RAW_disk[0].short_name[0]<s>, short_name<md100>) == <0>)
j++;
strcpy(short_name<md100>, GLOBAL_disk_info[76].short_name<(nil)>)
Segmentation Fault
---
 
The disks uses the SAN stuff look very different than they used to (world-wide names), which may be the problem.
Here is "format" output:
---
root at pf5d9:/usr/local/orca/bin> format
Searching for disks...done
 


AVAILABLE DISK SELECTIONS:
       0. c11t60020F200000EF2D4071BF2A00073353d0 <SUN-T300-0201 cyl 58523 alt 2 hd 14 sec 128>
          /scsi_vhci/ssd at g60020f200000ef2d4071bf2a00073353
       1. c11t60020F200000EF2D40688A1A000723D8d0 <SUN-T300-0201 cyl 42136 alt 2 hd 7 sec 128>
          /scsi_vhci/ssd at g60020f200000ef2d40688a1a000723d8
       2. c11t60020F200000EF2D407189BB00011A5Ed0 <SUN-T300-0201 cyl 58523 alt 2 hd 14 sec 128>
          /scsi_vhci/ssd at g60020f200000ef2d407189bb00011a5e
       3. c11t60020F200000F5CD4073C175000D429Ed0 <SUN-T300-0201 cyl 23408 alt 2 hd 7 sec 128>
          /scsi_vhci/ssd at g60020f200000f5cd4073c175000d429e
       4. c11t60020F200000F5CD40441C8F00046208d0 <SUN-T300-0201 cyl 42136 alt 2 hd 7 sec 128>
          /scsi_vhci/ssd at g60020f200000f5cd40441c8f00046208
       5. c11t60020F200000F5CD40718FAF000EF21Cd0 <SUN-T300-0201 cyl 28090 alt 2 hd 7 sec 128>
          /scsi_vhci/ssd at g60020f200000f5cd40718faf000ef21c
       6. c11t60020F200000F5CD40718FEC000E2308d0 <SUN-T300-0201 cyl 42136 alt 2 hd 7 sec 128>
          /scsi_vhci/ssd at g60020f200000f5cd40718fec000e2308
       7. c11t60020F200000F5CD407194CB0002FE4Cd0 <SUN-T300-0201 cyl 58523 alt 2 hd 14 sec 128>
          /scsi_vhci/ssd at g60020f200000f5cd407194cb0002fe4c
       8. c11t60020F200000F5CD4071876C000AC718d0 <SUN-T300-0201 cyl 46818 alt 2 hd 14 sec 128>
          /scsi_vhci/ssd at g60020f200000f5cd4071876c000ac718
---
 
Can you help me figure this out?  I'm not sure what to do.  
Thanks!
 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: </pipermail/orca-users/attachments/20040521/e4ccb273/attachment.html>